A closer look at adaptive regret
- Submitting institution
-
University of Brighton
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 7130560
- Type
- D - Journal article
- DOI
-
-
- Title of journal
- The Journal of Machine Learning Research
- Article number
- -
- First page
- 1
- Volume
- 17
- Issue
- 23
- ISSN
- 1532-4435
- Open access status
- Out of scope for open access requirements
- Month of publication
- April
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Merging forecasts in dynamic environments when neither the horizon (end time) nor the start time for forecasting are known in advance is an increasing challenge in many application areas. This paper is significant because it uses the specialist (sleeping experts) proof technique to obtain adaptive regret guarantees. This work has inspired theoretical developments in the field (Mohri and Yang, PMLR 2018) and applications to a range of domains, including forecasting house prices (Adamskiy et al., Mach Learn 2018), disease spread (Baby et al., arxiv:2101.09438) and electricity consumption (V’yugin and Trunov, Mach Learn 2019).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -